Sergey P Orlov

Bio: Sergey P Orlov is an academic researcher from Samara State Technical University. The author has contributed to research in topics: Artificial neural network & Petri net. The author has an hindex of 2, co-authored 17 publications receiving 21 citations.

Artificial Neural Network for Technical Diagnostics of Control Systems by Thermography

Abstract: The paper deals with the problem of testing and technical diagnostics of the control systems in electronic devices by means of infrared thermography. The methodology of technical state noncontact monitoring and failure analysis is presented. An algorithm for measuring the temperature field of the elements surface and supporting decision procedure on the operability implemented in the information-measuring system is described. The different mathematic models of electronic devices thermal performance are considered. It is shown that for microelectronic devices it is sufficient to use a two-dimensional mathematical model of the thermal field on the surface. To compare the calculated thermal fields and measured temperatures, it is suggested to use an artificial neural network that is trained in the process of functioning of the information-measuring system. The experiments were carried out on the recognition of defects and failures in microelectronic devices of control systems.

Intelligent Information Processing System for Monitoring Rail Tracks

Abstract: The article deals with the operational processing of visual information during the inspection of a railway track. The structure of an intelligent information system for monitoring rail fasteners is proposed. The system is installed in the detector car and contains an automated workstation, a diagnostic neural network, and four video cameras for receiving visual information. The problem of operational recognition of rail fastening system defects is solved by using a deep convolutional network. The procedure for preparing training and test datasets is described. The problem of a small number of defective rail fastener images was solved using the technique of synthesizing augmented images. The experiments with the intellectual information system and artificial neural network were conducted. Determination of the classification accuracy when system working in real-time at high speeds of the detector car has been carried out. A confusion matrix was used to assess the quality of the rail fastener classification.

Intelligent Technologies in the Diagnostics Using Object’s Visual Images

Abstract: The problem of complex industrial equipment diagnostics using images in different spectral ranges is considered. An intelligent method for technical states classification according to images of a control object is proposed. Considered a neural network analyzer designed as a two-branch neural network. Convolutional neural network processes simultaneously three object’s images obtained in the visual, ultraviolet and infrared bands. The properties of the dataset for learning the neural network are investigated using the dimensionality reduction methods. Examples of the developed method and the neural network analyzer application for monitoring various industrial facilities are given.

Application of Hierarchical Colored Petri Nets for Technological Facilities’ Maintenance Process Evaluation

Abstract: The high reliability of modern engineering systems is achieved by performing predictive maintenance. Mathematical models based on stochastic timed colored Petri nets are an effective tool for developing complex production processes for Industry 4.0. This article discusses the maintainability evaluation used in hierarchical Petri net models. The hierarchical simulation model was built using timed colored Petri nets, and was constructed with four levels of repair and maintenance modules. New module structures are proposed for simulating the schedule of production tasks and interaction with technological units. The emphasis is on the processes of predicting maintenance and repair, moving units to service, replacing units, and forming a reserve. The design of the simulation modules allows the setting of probabilistic parameters for the distributions of equipment failures, requests for unit maintenance, repair time, and recovery time after repair. The article proposes to use the hierarchical Petri model in conjunction with solving the problem of minimizing the cost of service. The iterative procedure consists of obtaining an approximate unit distribution by tasks, subsequent simulation of the technological process, and adjusting the optimization problem constraints. For example, the hierarchical Petri net is considered to assess the maintainability of autonomous agricultural vehicles. The results of the simulation experiments are presented. A simulation of the agrotechnical production process was performed, during which vehicles were maneuvered, taken out for repair or maintenance, and returned to the reserve fund. The interdependencies of preventive maintenance periods, service operations, failure rates, and predictive maintenance requests were obtained in order to comply with the task scheduling. The proposed model is a generalization, but it is especially effective in studying mobile equipment servicing.

Machine Vision System for Autonomous Agricultural Vehicle

Abstract: The problem of the vehicle motion control for automated driving is considered. The robotic chassis is designed for agriculture and should work in the absence of roads. The general structural diagram of the chassis’ machine vision system is given. A structural obstacle separation method for constructing obstacle maps on the ground is proposed. The method uses lidars to detect obstacles. The implementation of the method is based on the assumption that the terrain within the area no significant differences in elevation (gullies, dips) and water obstacles. For solving the detecting obstacle problem in these conditions, it is enough to detect points which height exceeds a certain threshold, and to identify the relationship between these points to assess the obstacle size. The clustering of points in each layer by the Euclidean distance is performed. Then, the coordinates of the cluster centers are recalculated into the global coordinate system, which allows transferring obstacles to an area map, taking into account the dimensions that determine the degree of the detected obstacle danger. An algorithm for implementing the proposed method is described. The report also provides information on the composition of the software for the robotic chassis machine vision. Simulation results of the proposed method of the allocation of obstacles are presented. The method has low computational complexity, which reduces the requirements for the robotic chassis on-board computer.

The World as a User Interface: Augmented Reality for Ubiquitous Computing.

Abstract: We discuss the possibilities of augmented reality (AR) as a ubiquitous user interface to the real world. A mobile AR system can constantly provide guidance to its user through visual annotation of the physical environment. The first part of the paper discusses the necessary ingredients for ubiquitous AR, on which we have worked in the recent past, namely mobile AR hardware, wide area tracking, unobtrusive user interfaces, application prototypes, and geographic data models suitable for AR. The second part of the paper examines future requirements of such data models in greater detail. Based on the lessons learned in our previous work, we identify shortcoming of existing standards for geographic information systems and visualization models. Ubiquitous AR requires independence of the data model from specific applications and their implicit assumptions. A semantic network model of geo-referenced data provides such a data model. We examine how such a model fits the requirements of AR applications, and how it can be implemented in practice.

A Model for Predictive Maintenance Based on Asset Administration Shell.

Abstract: Maintenance is one of the most important aspects in industrial and production environments. Predictive maintenance is an approach that aims to schedule maintenance tasks based on historical data in order to avoid machine failures and reduce the costs due to unnecessary maintenance actions. Approaches for the implementation of a maintenance solution often differ depending on the kind of data to be analyzed and on the techniques and models adopted for the failure forecasts and for maintenance decision-making. Nowadays, Industry 4.0 introduces a flexible and adaptable manufacturing concept to satisfy a market requiring an increasing demand for customization. The adoption of vendor-specific solutions for predictive maintenance and the heterogeneity of technologies adopted in the brownfield for the condition monitoring of machinery reduce the flexibility and interoperability required by Industry 4.0. In this paper a novel approach for the definition of a generic and technology-independent model for predictive maintenance is presented. Such model leverages on the concept of the Reference Architecture Model for Industry (RAMI) 4.0 Asset Administration Shell, as a means to achieve interoperability between different devices and to implement generic functionalities for predictive maintenance.

Artificial Neural Network for Technical Diagnostics of Control Systems by Thermography

Abstract: The paper deals with the problem of testing and technical diagnostics of the control systems in electronic devices by means of infrared thermography. The methodology of technical state noncontact monitoring and failure analysis is presented. An algorithm for measuring the temperature field of the elements surface and supporting decision procedure on the operability implemented in the information-measuring system is described. The different mathematic models of electronic devices thermal performance are considered. It is shown that for microelectronic devices it is sufficient to use a two-dimensional mathematical model of the thermal field on the surface. To compare the calculated thermal fields and measured temperatures, it is suggested to use an artificial neural network that is trained in the process of functioning of the information-measuring system. The experiments were carried out on the recognition of defects and failures in microelectronic devices of control systems.

Intelligent Information Processing System for Monitoring Rail Tracks

Abstract: The article deals with the operational processing of visual information during the inspection of a railway track. The structure of an intelligent information system for monitoring rail fasteners is proposed. The system is installed in the detector car and contains an automated workstation, a diagnostic neural network, and four video cameras for receiving visual information. The problem of operational recognition of rail fastening system defects is solved by using a deep convolutional network. The procedure for preparing training and test datasets is described. The problem of a small number of defective rail fastener images was solved using the technique of synthesizing augmented images. The experiments with the intellectual information system and artificial neural network were conducted. Determination of the classification accuracy when system working in real-time at high speeds of the detector car has been carried out. A confusion matrix was used to assess the quality of the rail fastener classification.

Optimization-simulation approach to the computational resource allocation in a mechanical engineering enterprise

Abstract: The complexity of modern engineering production requires the use of computer technology in design and technological processes. The problem is the need to efficiently distribute computers and software between departments in the enterprise. The article proposes an optimization-simulation approach to solving the problem of computer equipment allocation. It consists in an iterative procedure for solving an optimization problem and verifying the obtained solutions on simulation models. A discrete optimization problem with Boolean variables for assigning computers to tasks in departments is formulated. It is proposed to carry out a simulation model in the form of a timed Petri net. A graphical model of computer equipment repair and maintenance in the process of performing tasks is presented. A basic elementary timed Petri net is constructed, which is used to build a general simulation model. The properties of the basic simulation model are considered.